Nghiên cứu liên kết hydrogen kích hoạt phân ly khí nitric oxide trên bề mặt cu(110) bằng mô phỏng máy tính

ĐẠI HỌC QUỐC GIA TP. HCM
TRƯỜNG ĐẠI HỌC BÁCH KHOA
--------------------

PHẠM NGỌC THANH

NGHIÊN CỨU LIÊN KẾT HYDROGEN KÍCH HOẠT
PHÂN LY KHÍ NITRIC OXIDE TRÊN BỀ MẶT CU(110)
BẰNG MƠ PHỎNG MÁY TÍNH
Chun ngành: Khoa Học Tính Tốn
Mã số: 60460136

LUẬN VĂN THẠC SĨ

TP. HỒ CHÍ MINH, tháng 6 năm 2018


VIETNAM NATIONAL UNIVERSITY – HO CHI MINH CITY
UNIVERSITY OF TECHNOLOGY
--------------------

PHAM NGOC THANH

STUDY HYDROGEN BOND INDUCED NO
DISSOCIATION ON CU(110) BY COMPUTER
SIMULATION
Major: Computational Science
ID: 60460136

MASTER THESIS

HO CHI MINH CITY, June 2018


CƠNG TRÌNH ĐƯỢC HỒN THÀNH TẠI
TRƯỜNG ĐẠI HỌC BÁCH KHOA –ĐHQG -HCM
Cán bộ hướng dẫn khoa học 1: Prof. Yoshitada Morikawa .................................
(Ghi rõ họ, tên, học hàm, học vị và chữ ký)
Cán bộ hướng dẫn khoa học 2: TS. Đỗ Ngọc Sơn ................................................
(Ghi rõ họ, tên, học hàm, học vị và chữ ký)
Cán bộ chấm nhận xét 1 : .....................................................................................
(Ghi rõ họ, tên, học hàm, học vị và chữ ký)
Cán bộ chấm nhận xét 2 : .....................................................................................
(Ghi rõ họ, tên, học hàm, học vị và chữ ký)
Luận văn thạc sĩ được bảo vệ tại Trường Đại học Bách Khoa, ĐHQG Tp. HCM
ngày . . . . . tháng . . . . năm . . . . .
Thành phần Hội đồng đánh giá luận văn thạc sĩ gồm:
(Ghi rõ họ, tên, học hàm, học vị của Hội đồng chấm bảo vệ luận văn thạc sĩ)
1. ....................................................................
2. ....................................................................
3. ....................................................................
4. ....................................................................
5. ....................................................................
Xác nhận của Chủ tịch Hội đồng đánh giá LV và Trưởng Khoa quản lý chuyên
ngành sau khi luận văn đã được sửa chữa (nếu có).
CHỦ TỊCH HỘI ĐỒNG

TRƯỞNG KHOA…………

i

ĐẠI HỌC QUỐC GIA TP.HCM
TRƯỜNG ĐẠI HỌC BÁCH KHOA

CỘNG HÒA XÃ HỘI CHỦ NGHĨA VIỆT NAM
Độc lập - Tự do - Hạnh phúc

NHIỆM VỤ LUẬN VĂN THẠC SĨ
Họ tên học viên: PHẠM NGỌC THANH

MSHV:1670060

Ngày, tháng, năm sinh: 20/08/1993

Nơi sinh: Gia Lai

Chun ngành: Khoa học tính tốn

Mã số: 60460136

I. TÊN ĐỀ TÀI: Nghiên cứu liên kết hydrogen kích hoạt phân ly khí nitric oxide
trên bề mặt Cu(110) bằng mơ phỏng máy tính (Study hydrogen bond induced NO
dissociation on Cu(110) surface by computer simulation).
II. NHIỆM VỤ VÀ NỘI DUNG:
Học viên nghiên cứu đánh giá việc hấp phụ và phân ly khí NO trên bề mặt
Cu(110). Đánh giá khả năng kích hoạt phân ly NO bằng liên kết hydrogen thơng qua
việc tính tốn đường phản ứng có năng lượng kích hoạt nhỏ nhất trong các trường
hợp phân tách NO có và khơng có mặt liên kết hydrogen. Cuối cùng, tính tốn cấu
trúc điện tử dẫn đến việc kích hoạt phân ly NO.
The molecular and dissociative adsorptions of NO on Cu(110) are elucidated. The

hydrogen bond-induced NO dissociation on Cu(110) is investigated by calculating
the minimum energy pathways for NO dissociation with and without hydrogen bond
coupling. The electronic orgins leading to the induction effect are proposed through
the electronic structure calculations.
III. NGÀY GIAO NHIỆM VỤ : 15/01/2018
IV. NGÀY HOÀN THÀNH NHIỆM VỤ: 02/12/2018
V. CÁN BỘ HƯỚNG DẪN : Prof. Yoshitada Morikawa và TS. Đỗ Ngọc Sơn.

Tp. HCM, ngày . . . . tháng .. . . năm 20....
CÁN BỘ HƯỚNG DẪN

CHỦ NHIỆM BỘ MÔN ĐÀO TẠO

(Họ tên và chữ ký)

(Họ tên và chữ ký)

TRƯỞNG KHOA KHOA HỌC ỨNG DỤNG

(Họ tên và chữ ký)

ii


Acknowledgements
First, I would like to thank my supervisors, Prof. Yoshitada Morikawa (Graduate School
of Engineering, Osaka University (OU)) and Dr. Do Ngoc Son (Ho Chi Minh city University
of Technology (HCMUT)) for their guidance, encourages, helpful comments throughout my
master course.
I must say thank to Prof. Yoshitada Morikawa for providing me a chance to conduct my

research at his laboratory. I have been extremely lucky to have a supervisor who cared so
much about my work, who responded to my questions and queries so promptly. I admire his
huge efforts to take care of my research every week, guide me to study machine learning
techniques during my first three months at OU. I really enjoy our every Saturday meeting
during my time at OU so much. Furthermore, I am thankful for the doctor course opportunity
at OU from him.
I am grateful to express my deepest appreciation for the guidance of Dr. Do Ngoc Son
when I studied and conducted the research at HCMUT. His enthusiasm for research has a
huge impact to me and inspires me to become an academic researcher. In addition, I would
like to thank his recommendation for my research at OU.
During the time when I conducted my research at OU, I am grateful to have fruitful
discussions, insightful comments, and kind helps from M.Sc. Mashahiro Sugiyama, M.Sc.
Setia Eka Marsha Putra, Dr. Fahdzi Muttaqien, Prof. Kouji Inagaki, Prof. Yuji Hamamoto,
and Prof. Ikutaro Hamada. The helpful comments on my manuscript made by Prof. Ikutaro
Hamada are highly appreciated. In addition, I would like to thank all of the students and
staffs at Morikawa Lab for their helps and I really enjoy my great time at OU.
I am grateful for financial supports by Japan Student Services Organization (JASSO) and
Toshiba scholarship for VNU-HCM. In addition, the part-time research assistant job from
Prof. Morikawa are highly appreciated.
Finally, I would like to thank the supports from my family and my girlfriend.

iii


Tóm tắt luận văn
Trong luận văn thạc sỹ này, học viên nghiên cứu phân ly của khí nitric oxide (NO)
trên bề mặt Cu(110) và ảnh hưởng của liên kết hydrogen lên phân ly NO bằng lý
thuyết phiếm hàm mật độ. Học viên tìm thấy rằng khí NO có thể hấp phụ phân tử bền
vững trên bề mặt Cu(110) dưới dạng cấu hình thẳng đứng tại vị trí cầu nối ngắn và ở
dưới dạng cấu hình nằm ngang tại vị trí rỗng kết nối hai vị trí cấu nối ngắn gần nhau.

Dạng cấu hình nằm ngang là trạng thái trung gian cho quá trình phân ly NO. Các
đường phản ứng phân ly NO trên Cu(110) dưới sự ảnh hưởng của liên kết hydrogen
với phân tử nước được nghiên cứu hệ thống. Học viên tìm được rằng liên kết hydrogen
có thể kích hoạt phân ly NO với hàng rào năng lượng cực thấp khi có sự hình thành
một liên kết hydrogen giữa một dime nước và nguyên tử O của NO nằm ngang và hai
liên kết hydrogen giữa mỗi dime nước với mỗi đầu của NO nằm ngang. Quan trọng
hơn, hiệu ứng thúc đẩy phân ly NO chỉ hiệu quả khi có sự hình thành liên kết hydrogen
giữa một dime nước và phân tử O của NO nằm ngang. Kết quả luận văn này cung cấp
cơ chế về mặt vật lý cho hiệu ứng kích hoạt phân ly NO bằng liên kết hydrogen với
nước, mà có thể hữu dụng trong việc cải thiện hiệu năng xúc tác và thiết kế các chất
xúc tác mới trong lĩnh vực giảm thiểu phát xạ khí NO ra khơng khí.

iv


Abstract
In this master thesis, I have studied the dissociation process of nitric oxide (NO)
on Cu(110) and the influence of the hydrogen bond with water by means of density
functional theory calculations. I have found that an upright NO adsorbed at a shortbridge site and a side-on NO at a hollow site connecting two short-bridge sites are the
two most stable molecularly adsorbed states, and the latter is the precursor for the
dissociation process. Various NO dissociation pathways under the influences of the
hydrogen bonds with water have been investigated. I have found that hydrogen bonds
efficiently reduce the activation energy of NO dissociation by the introductions of a
water dimer to O and water dimers to both sides of the side-on NO, respectively.
More importantly, the promoting effect of water molecules on NO dissociation is
dominant only when one of water molecules in a water dimer forms a hydrogen bond
with O of the side-on NO. My results provide a physical insight into the promoting
effect of hydrogen bonds with water, which may be helpful in improving catalytic
activity as well as designing novel catalysts for NO reduction.

v


Declaration
I declared that this thesis was composed by myself, that the work herein is my
own except where explicitly stated otherwise in the text, and this work has not been
submitted for any other degree or processional qualification excepts as specified.

Parts of this work have been published in T.N. Pham, M. Sugiyama, F. Muttaqien,
S. E. M. Putra, K. Inagaki, D. N. Son, Y. Hamamoto, I. Hamada, Y. Morikawa,
“Hydrogen Bond-Induced Nitric Oxide Dissociation on Cu(110)”, J. Phys. Chem. C,
122, 11814 (2018).

vi


List of abbreviations

No.

Abbreviation

Definition

1

TWC

Three-way catalyst

2

NO

Nitric oxide

3

CO

Carbon monooxide

4

HC

Hydrocarbon

5

RAIRS

Reflection absorption infrared spectroscopy

6

EELS

Electron energy loss spectroscopy

7

LEED

Low energy electron diffraction

8

STM

Scanning tunneling microscopy

9

DFT

Density functional theory

10

GGA

Generalized gradient approximation

11

ML

Monolayer

12

PDOS

Projected density of states

13

vdW-DF

Van der Waals density functionals method

14

MEP

Minimum energy pathway

15

LDA

Local density approximation

vii


Generalized direct inversion of iterative

16

GDIIS

17

PBE

Perdew-Burke-Ernzerhof

18

STATE

Simulation tool for atom technology

19

NEB

Nudged elastic band

20

CI-NEB

Climbing image nudged elastic band

subspace

viii

List of figures
Figure
1.1

Caption
System integrating exhaust gas recirculation [5].

Page
1

The NO dimerization on Cu(110) visualized by STM techniques.
1.2

One of NO in an upright configuration are manipulated to near

4

other upright NO to form dimer (NO)2 on Cu(110) [38].
1.3

1.4

NO monomer configurations and their N-O bond stretching modes
[12, 13, 45].
Schematic illustrations of π back donation upon NO adsorption on
the Pt surface [12].

4

5

The left panel: calculated adsorption energies for NO adsorption
on the metal surfaces. Asterisk denotes the value of experimental
1.5

heat of adsorption. The right panel: The correlation between N–O

6

bond length (𝑑#$% ) and their stretching modes (𝜈#$% ). In gas
phase, 𝑑#$% = 1.168Åand 𝜈#$% = 1917cm-1 [21,22].
1.6

1.7

1.8
1.10

Mechanism of NO dissociation on flat metal (111) surfaces
[21,22].
Potential energy surface of NO dissociation on metal surface
[21,22].
Minimum energy pathway (MEP) for NO dissociation on Rh(111)
and Rh(211) [26].
Potential energy diagram of NO/Cu(110) [37].
ix

7

7

9
10


Projected density of states (PDOS) of the upright NO monomer at
1.11

short-bridge site. The strong hybridizations between 2π* orbitals

10

and d band of Cu lead to the splitting of 2π* peaks in PDOS [47].
(a) an STM image of NO and H2O on the Cu(110) surface, (b) an
STM image of NO···HOH complex obtained by manipulating the
1.12

water molecule toward adsorbed NO. (c), and (d) are schematic

12

illustration of adsorbates. The lines show the Cu(110) lattices.
Color: N, blue; O, red; H, white [35].
1.13

STM images of the products yielded by the reaction of NO···HOH
with another water molecule [35].

13

Schemiatic illustration for all-electron (solid lines) and pseudo
2.1

wave potentials (dashed lines) and their corresponding wave

21

functions [52, 53].
(a): Adsorption sites for the upright N*O on the Cu(110) p(2× 3)
surface: on-top (T), long-bridge (LB), short-bridge (SB), and
hollow (H); (b) Schematic diagram of MEP for NO dissociation.
3.1

The most stable short-bridge upright, metastable precursor,

27

transition, and final product states are denoted by SB-N*O, N*O*,
TS1, and N*+ O*, respectively. Color schemes: Cu: orange, O:
red, N: blue.
Top view of most favorable water monomer adsorption and the
3.2

N*O···HO*H complexes on the Cu(110) surface. Color scheme:

31

Orange: Cu, Blue: N, Red: O, White: H.

The ZPE corrected MEP for NO dissociation starting with
3.3

N*O···HO*H complexes on the Cu(110) p(2× 3). Color schemes:
Cu: orange, O: red, N: blue.
x

33


The ZPE corrected MEP for NO dissociation starting with H2O*+
3.4

N*O···HO*H on Cu(110). Color schemes: Cu: orange, O: red, N:

34

blue.
The ZPE corrected MEPs for NO dissociation on Cu(110) starting
3.5

with HO*H···OH2···O*N* (red line) and HO*H···OH2···O*N* (

36

blue line). Color schemes: Cu: orange, O: red, N: blue, H: White.
The ZPE corrected MEP for NO dissociation on the Cu(110) p(2
3.6

×5) surface starting with 2HO*H···OH2···N*O*. Red line (blue

line) denotes dissociative (associative) pathway. Color schemes:

38

Cu: orange, O: red, N: blue, H: White.
Projected densities of states (PDOSs) onto the molecular orbitals
of the isolated NO molecule. The upper part demonstrates the
3.7

valence states of NO (the occupied 1π, 5σ orbitals, and
antibonding partially occupied 2π* orbitals) with four topmost Cu

40

atoms. Doubly degenerate π[223] ,π[332] of π orbitals are dumbbell
shapes along [001] and [110], respectively
Charge

density

differences

HO*H···OH2···O*N*

with

for
Cu

N*O*

+

surface

H2O*

(a),

(b),

and

HO*H···OH2···O*N* without Cu surface (c). The atomic
3.8

geometry in the HO*H···OH2···O*N* without Cu surface is kept
fixed with that in the presence of substrate. Yellow and cyan
indicate charge accumulation and depletion, respectively. The
isosurface is plotted at 0.0067 e$ ×Å$7

xi

41


List of tables
Table

Caption

Page

The dissociation barrier (𝐸9:;; ), transition state energy (𝐸<= ),
1.1

rebonding energy (𝐸>?@AB9 ) and intramolecular N-O interaction
energy (𝐸:BC,<= ) at transition states of NO dissociation on several

8

surfaces calculated by PW91 functional [21, 22].
#%
Adsorption energy of single NO molecule (𝐸E9;
), N-O bond

3.1

length (𝑑#$% ), height of N of NO from the topmost Cu layer
(ℎ#$GH ), tilting angle of the adsorbed NO from the surface

28

normal (∠#%$GH ).a
Calculated energy difference between SB-N*O and N*O* (ΔE),
energy barrier for the flipping process from SB-N*O to N*O*
KLM

3.2

(𝐸E ), activation energy (𝐸E9:;; ), and effective activation energy
9:;;
(𝐸?KK
) of NO dissociation on the Cu(110) surface. The ZPE

30

corrected energies are in parentheses. The positive (negative)
value of ΔE indicates SB-N*O is more (less) stable than N*O*.
#%OP %

Calculated ZPE corrected co-adsorption energies (𝐸NA$E9;Q ),
3.3

#%OPQ %

hydrogen bond energy (𝐸PR

), N-O bond length (𝑑#$% ) of

32

N*O···HO*H complexes.
Calculated ZPE corrected co-adsorption energy (𝐸NA$E9; ), N-O
3.4

bond length (𝑑#∗ $% ), and ZPE corrected hydrogen bond energy
(𝐸PR ) at initial molecularly co-adsorbed states. The activation
KLM

energy for the flipping process of NO (𝐸E ), the activation

xii

35


9:;;
energy (𝐸E9:;; ), and the effective activation energy (𝐸?KK
), along

with the reaction energy (E), are also included.
The calculated ZPE corrected co-adsorption energy (𝐸NA$E9; ),
the N*-O* bond length (𝑑#∗ $%∗ ), the hydrogen bond length
3.5

between N*O* and water dimer (𝑑#∗ %∗ $P ), and the internal
hydrogen bond length of the water dimer (𝑑P%∗ P$%PQ ) at initial
states. The activation energy (𝐸E ) and the reaction energy (E)
for the NO dissociation are presented.

xiii

37


Contents
1. Introduction

1

1.1

NO reduction in three-way catalysts

1

1.2

NO adsorption on metal surfaces

2

1.2.1. Molecularly adsorbed NO on metal surfaces

4

1.2.2. NO dissociation on the metal surfaces

6

1.3

NO adsorption and dissociation on the Cu surfaces

10

1.4

Hydrogen bond-induced NO dissociation on Cu(110)

11

1.5

Research objects and Thesis outline

13

2. Methodology

16

2.1

Many-body problem

16

2.2

Hohenberg-Kohn theorems

17

2.3

Kohn-Sham equation

18

2.4

Exchange-correlation functional

19

2.5

Bloch’s theorem and plane wave basis sets

20

2.6

Pseudopotentials

21

2.7

Self-consistent field

22

2.8

Van der Waals density functionals (vdW-DFs) method

23

2.9

Computational details

24
xiv


3

4

Results and discussion

26

3.1

Dissociative adsorption of NO on the Cu(110) surface

26

3.2

Formation of N*O···HO*H complex

31

3.3

Effect of water monomers on NO dissociation

33

3.4

Effect of water dimer in NO dissociation

36

3.5

Electronic structure analysis

39

Conclusions and outlooks

43

4.1

Conclusions

43

4.2

Outlooks

43

5

Reference

45

6

Appendix

S1

Curriculum vitae

S4

xv


Chapter 1. Introduction
1.1 . NO reduction in three-way catalysts
Up to now, mankind has been facing with numerous environmental issues relating
to atmospheric pollutions from the automotive operations such as acid rain, global
warming, and ozone depletion [1–9]. The use of energy fuels (diesel, gasoline, etc.)
produces a large number of the exhaust gas consisting of nitric oxide (NO), carbon
monoxide (CO), and un-combusted hydrocarbons (HCs), which is directly emitted to
the atmosphere. To control the exhaust gas emission, the implementation of aftertreatment systems provided remarkable results in the past three decades. The harmful

gases (NO, CO, and HCs) are converted into harmless substances (N2, CO2, and H2O).
The schematic of the after-treatment system is depicted in Figure 1.1.

Figure 1.1. Schematic of after-treatment systems [5].
Among numerous available technological solutions for the after-treatment
systems, three-way catalysts (TWCs) correspond to a first breakthrough owing to
their excellent conversions of higher than 95% for CO, unburned HCs, and NOx [1–
9]. Currently, TWCs are optimized to promote the following oxidation and reduction
processes of exhaust gas occur simultaneously.
Oxidation.
$

CO + O% → CO% .
%

1


C' H ) + 𝑥 +

)
+

)

O% → 𝑥CO% + H% O.
%

Reduction.
NO + H% →

1
N + H% O.
2 %
$

NO + CO → N% + CO% .
%

2𝑥 +

)
%

NO + C' H) → 𝑥 +

)
+

)

N% + 𝑥CO% + H% O.
%

The supported noble metal catalysts consisting of the Pt, Rh, and Pd nanoparticles
dispersed on metal oxides support Al2O3 or SiO2 promoted by CeO2 are widely
employed in TWCs. The combination of Pt and Rh provides a complete removal of
NO, CO, and HCs simultaneously, whereas Pd is adopted here to replace Pt since it
is less expensive [1–9].
In those reduction processes, the NO reduction directly to N2 is desirable [1–28];

however, the production of a side product (N2O) especially in a low temperature
regime as well as in a lean-burn condition lowers the NO conversions, being a serious
drawback for exhaust gas elimination. So far, Rh has been emerged as the most
effective metal for the reduction of NO because NO is easily adsorbed and cleaved
as well as the N2O production is low. Nevertheless, Rh is one of the rarest and the
most expensive metals, hence Rh replacements are inevitable. Alternative noble
metals (Ru, Ir, Au, Ag, Cu, as well as alloys of those metals with cheaper metals)
dispersed on a support have been evaluated for NO reduction but most of the attempts
for Rh replacements failed [1–28]. Thus, finding novel catalysts that can replace Rh,
is a crucial task for academia and industry.

1.2. NO adsorption on metal surfaces
For NO reduction with CO, HCs, or H2, the detailed reaction mechanism varies
regarding to metal surface, gas pressure, as well as nature of support materials.
Nevertheless, there is a strong agreement that molecular NO adsorption and

2


dissociative NO adsorption to N and O adatoms are important elementary steps since
the NO reduction cannot proceed unless the N-O bond is cleaved [1–28].
To understand the mechanism of the conventional TWC, NO adsorption and the
reactivity of NO on metal surfaces have been investigated intensively for several
decades. In contrast with abundant experimental and theoretical studies on CO
adsorbed on metal surfaces [29–34], it appears that the fundamentals of NO on
surfaces are poorly understood owing to high chemical reactivity of NO. On metal
surfaces, NO can form various adsorbate species, e.g., NO, N, O, (NO)2, N2O, and
NO2 depending on NO coverage and the nature of metal surfaces, leading to the
challenge for the experimental characterization and identification of individual
species [12, 13, 21, 22]. Brown et al. [12] reviewed the comprehensive NO reactivity

on metal surfaces and they proposed a tendency of the NO reactivity on metals
surfaces at low coverages regimes in Figure 1.2.
Vibrational frequency measurements of the adsorbed NO are fingerprints to trace
down molecular structure of the NO adsorption including preferred adsorption sites,
axial geometries, etc., being important considerations to elucidate the NO reactivity
at metal surfaces [13]. The experimental techniques [35–44] for vibrational studies
are reflection absorption infrared spectroscopy (RAIRS), electron energy loss
spectroscopy (EELS), and low energy electron diffraction (LEED). At low
temperature, scanning tunneling microscopy (STM) techniques are powerful
techniques that provide a direct time-dependent observation for NO on the metal
surfaces to characterize the not only the configurations of the adsorbed NO but also
the NO reaction pathways on the metal surfaces, i.e. initial, intermediate, and product
states. Recently, by adopting STM techniques [38], a direct observation of NO
dimerization on Cu(110) (2NO → (NO)2) are obtained, see Figure 1.3. In addition,
theoretical studies replied on density functional theory (DFT) are now in use to study
the atomic configurations of NO adsorption states, stable adsorption sites; as well as
elucidate the interaction of the NO on the metal surfaces including bonding and

3


electronic structures, providing the direct comparisons to verify the experimental
observations [19–28].

Figure 1.3. The NO dimerization on Cu(110) visualized by STM techniques. One
of NO in an upright configuration are manipulated to near other upright NO to form
dimer (NO)2 on Cu(110) [38].
1.2.1. Molecularly adsorbed NO on metal surfaces.

Figure 1.4. NO monomer configurations and their N-O bond stretching modes [12, 13,

45].

As the first step for NO reduction, the molecular adsorptions of NO on the metal
surfaces are under intensive investigations experimentally [11–13, 35–45] and
theoretically [19–28]. On the metal surfaces, NO can form numerous adsorbate
species such as NO monomer, dimer, and trimer depending on NO coverage on metal
4


surfaces. At low coverage regime, the NO monomer is dominant and NO is
molecularly adsorbed with several configurations, i.e. upright, bent, and flat-lying
ones depending on the tilting angle of NO respected to metal surface normal.
Generally speaking, preferred adsorption sites of the upright NO are highly
symmetric coordinated sites: atop, bridge, 3-fold hollow, and 4-fold hollow, see
Figure 1.4, [12, 13, 45]. Intensive reviews of molecular NO adsorption relating to
adsorption sites and their N-O bond stretching modes on the several metal surfaces
can be found in Ref [13].

Figure 1.5. Schematic illustrations of π back donation upon NO adsorption on the Pt
surface [12].

It is useful to provide an electronic origin upon the NO adsorption on the metal
surfaces. The DFT calculations have been widely used to verify the experimental
observations of NO adsorption and understand the bonding mechanism between NO
and metal surfaces [19–28]. In particular, DFT calculations with the generalized
gradient approximation (GGA) [46] and slab models for the metal substrate have been
produced remarkably good agreements in the atomic structures, vibrational
frequencies as well as bonding mechanisms upon NO adsorption [19–28]. The
bonding mechanisms of NO on metal surfaces are attributed to the π back donation,
see Figure 1.5, [12]. NO in gas phase has open shell structure in which 2π* orbitals

are half-filling. When NO is adsorbed on the metal surfaces, a strong coupling
5


between 2π* orbitals of NO and d band of the metal surfaces take place, leading to a
typically loss of unpaired electron. Nevertheless, depending on the origin of
interaction of NO and metal surfaces, open shell structure of NO on metal surface
can be retained, e.g. NO/Ag, and NO/Cu, as confirmed by the dimerization of the
adsorbed NO [38–40].

Figure 1.6. The left panel: calculated adsorption energies for NO adsorption on the
metal surfaces. Asterisk denotes the value of experimental heat of adsorption. The right
panel: The correlation between N–O bond length (𝑑345 ) and their stretching modes
(𝜈345 ). In gas phase, 𝑑345 = 1.168Åand 𝜈345 = 1917cm-1 [21,22].

Eichler et al. studied molecular adsorption of NO on several close-packed surfaces
of late transition metals (Co, Ni, Ru, Pd, Ir, and Pt) as well as noble metals (Cu, Ag,
Au) at a coverage of 0.25 monolayer (ML) within GGA-DFT [21,22]. The preferred
site of NO is the hollow site and NO adsorbs in an upright configuration with N binds
to the metal surface, expected for NO/Ir(111) and NO/Au(111) where NO adsorbed
at atop and bridge sites, respectively. Furthermore, trends in the NO adsorption and
stretching modes of the adsorbed NO on serval metal (111) surface are established as
shown in Figure 1.6. Their results provide a comprehensive understanding of NO
adsorption on the metal (111) surfaces.
1.2.2. NO dissociation on the metal surfaces.

As an important elementary step for NO reduction, NO dissociation is studied
experimentally [35–45] and theoretically [20–28]. The common mechanism for NO
dissociation into N and O adatoms on the metal (111) surfaces are shown in Figure

6


1.7, [21,22]. Initially, NO is adsorbed in upright configuration with N binds to the
metal surface. Then, NO is tilted together with the N-O bond is gradually elongated
until it orients almost parallel to the surface at the transition state. At this state, the
N-O bond is cleaved, yielding N and O adatoms on the surface. The stability of the
flat-lying configuration of NO at the transition state attributes directly to the reactivity
of the NO dissociation in which the more stable the transition state is, the lower
activation energy for NO dissociation is required. To characterize the stability of flatlying NO, the transition state energy (𝐸>? ) proposed by Hammer [21–23] are employed
(see Figure 1.8).

Figure 1.7. Mechanism of NO dissociation on flat metal (111) surfaces [21,22].

Figure 1.8. Potential energy surface of NO dissociation on metal surface [21,22].
𝐸>? measures the stability of flat-lying NO on metal surfaces respected to NO in
gas phase. We can decompose 𝐸>? into a rebonding energy (𝐸ABCDEF ) and an
7


intramolecular N-O interaction energy (𝐸GEH,>? ), which depict the co-chemisorption
strengths and the lateral repulsive interactions of N and O adatoms at the transition
state, respectively. Stability of flat-lying NO is a compromise between cochemisorption and lateral repulsive interactions of N and O, which the former one
favors the stability of flat-lying NO, whereas the latter one un-favors the formation
of this adsorbed configuration (see Table 1.1) [21–22].
Table 1.1. The dissociation barrier (𝐸FGJJ ), transition state energy (𝐸>? ),
rebonding energy (𝐸ABCDEF ) and intramolecular N-O interaction energy (𝐸GEH,>? ) at
transition states of NO dissociation on several surfaces calculated by PW91
functional [21, 22].
Metal

𝑬𝐝𝐢𝐬𝐬 /eV

𝑬𝐓𝐒 /eV

𝑬𝐫𝐞𝐛𝐨𝐧𝐝 /eV

𝑬𝐢𝐧𝐭,𝐓𝐒 /eV

Co

0.73

-1.67

-3.99

2.32

Ni

1.21

-2.61

-3.44

2.14

Cu

1.68

-1.22

-1.13

1.59

Ru

0.75

-2.65

-4.56

2.66

Rh

1.52

-2.58

-3.52

2.46

Pd

2.77

-2.59

-2.22

2.39

Ag

2.95

-0.43

1.69

0.83

Ir

1.46

-2.10

-2.90

2.26

Pt

2.34

-2.07

1.58

1.85

Au

3.08

-0.38

1.82

0.88

The atomic configurations of metal surfaces also have significant impacts on NO
reactivity [19,20, 26–28]. Generally speaking, step surfaces usually promotes the NO
8